Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 27.
Published in final edited form as: Psychol Addict Behav. 2025 Jan 27;39(8):689–701. doi: 10.1037/adb0001059

Integrating the Confluence Model and I3 Model to Predict Sexual Assault Perpetration Intentions

Mitchell Kirwan 1, Olivia Westemeier 2, Julia F Hammett 3, Cynthia A Stappenbeck 2, Kelly Cue Davis 3
PMCID: PMC12301992  NIHMSID: NIHMS2042715  PMID: 39869700

Abstract

Objective:

Sexual assault perpetration is widespread among young men. According to the Confluence Model, hostile masculinity and impersonal sex are trait-level factors associated with sexual assault perpetration likelihood. Additionally, state-level factors, including alcohol intoxication, current emotions, and ability to modulate one’s emotions, have been tied to sexual assault perpetration via the I3 Model. This study integrates these trait- and state-level factors into a single model to enhance its predictive power and better inform future interventions.

Methods:

Data were collected from 2019–2023. Young, single, non-problem drinking men, who had been sexually active with a woman within the past month (N=282; 34.8% people of color), completed background questionnaires and were randomly assigned to a positive or negative mood induction, and then to consume alcohol (target peak BrAC = .08%) or a control beverage before projecting themselves into a hypothetical, sexual scenario. Then, men indicated their likelihood of engaging in nonconsensual sexual activity with their hypothetical partner (i.e., perpetration likelihood).

Results:

A moderated-mediation structural equation model demonstrated that trait hostile masculinity and impersonal sex interacted with state alcohol intoxication and mood to predict state sexual arousal. State sexual arousal subsequently interacted with state impulsivity to predict state difficulties modulating emotions, which predicted perpetration likelihood during the hypothetical scenario.

Conclusions:

Integrating trait- and state-level factors is of paramount importance to understanding sexual assault prevention. Interventions targeting emotional modulation during sexual situations may be especially useful among aroused or impulsive individuals, due to their mood, alcohol intoxication, and attitudes regarding hostile masculinity and impersonal sex.

Introduction

Sexual assault perpetration, or engaging in sexual activity with a partner who is unable or unwilling to consent, is widespread in the United States. About 30–40% of young men self-report having previously perpetrated sexual assault (Anderson et al., 2021; Trottier et al., 2021). The Confluence Model is a prominent theoretical model of trait-level sexual assault predictors (e.g., hostile masculinity, impersonal sex; Malamuth et al., 1991). However, this model does not consider the contributions of state-level factors on men’s perpetration (Bruera et al., 2023). Consistent with the I3 Model (Slotter & Finkel, 2011), alcohol intoxication (George et al., 2024), state emotion regulation (Kirwan, 2023; Kirwan & Davis, 2023; Neilson et al., 2022), and specific state emotions (Davis et al., 2020; Kirwan, 2023; Kirwan, Davis et al., 2022a) are all state-level factors which either impel or (dis)inhibit sexual assault perpetration. Thus, this study seeks to gain a more comprehensive understanding of the trait- and state-level factors underlying men’s sexual assault perpetration by examining the ways in which trait-level Confluence Model predictors (hostile masculinity and impersonal sex) may interact with state-level I3 Model factors, such as acute alcohol intoxication and positive or negative affective experiences, to influence men’s state-level experiences of specific emotions (sexual arousal and impulsivity), state emotion regulation ability, and intentions to perpetrate sexual assault during a hypothetical sexual scenario.

The Confluence Model

The Confluence Model is a widely supported framework of trait-level sexual assault predictors among men, which has been empirically supported across cross-sectional and longitudinal studies (Bruera et al., 2023). The Confluence Model variables which are most strongly and proximally associated with perpetration are hostile masculinity and impersonal sex. Hostile masculinity is a personality profile comprised of elevated levels of insecurities, defensiveness, or distrustfulness towards women, which manifest themselves such that they reinforce hegemonic masculine ideals (Ray & Parkhill, 2023; Reidy et al., 2022). Specifically, hostile masculinity is indicated by factors such as hostility towards women, harassment of women, and adversarial heterosexual beliefs (Ray & Parkhill, 2023; Reidy et al., 2022; Seto, 2019). Impersonal sex refers to one’s tendencies to maintain a detached approach to sexual relationships and to have more positive attitudes about casual sex (Davis, Stewart et al., 2023), which is evidenced by having more casual sex partners and a higher number of sex partners overall (Bruera et al., 2023; Malamuth et al., 2021). Furthermore, the Confluence Model argues that the “confluence” of hostile masculinity and impersonal sex is what is most predictive of sexual assault perpetration, such that men who report high levels of both of these factors are at the greatest risk of perpetration, especially relative to men who possess high levels of only one or none of these factors (Bruera et al., 2023; Malamuth & Hald, 2017).

I3 Model

Although these Confluence Model constructs are key trait-level contributors to sexual assault perpetration, the Confluence Model does not account for state or situational factors that may also influence perpetration. In contrast, the I3 Model describes three main components underlying the in-the-moment perpetration of aggression (including sexual assault): instigation, impellance, and (dis)inhibition (Slotter & Finkel, 2011). Instigators are situational “triggers” which induce perpetration. Impellors refer to trait- or state-level factors which increase an individual’s urges or readiness to perpetrate. Finally, inhibitors refer to trait- or state-level factors that increase an individual’s ability to resist an urge to perpetrate (Finkel et al., 2012; Slotter & Finkel, 2011). Similarly, disinhibitors, such as alcohol intoxication, reduce an individual’s ability to override an urge to aggress (Finkel, 2014).

Emotions, Emotion Regulation, and Alcohol Intoxication

Emotion regulation difficulties refer to the inability to regulate the experience and expression of emotions (Gross, 1998). Within the I3 Model, emotion regulation difficulties disinhibit sexual assault perpetration among men (Kirwan, Lanni et al., 2019; Kirwan, Svenson et al., 2019; Pickett et al., 2016). Men who have difficulties modulating emotions, including impulsivity (feeling the need to act compulsively, without reason or deliberation; Evenden, 1999) and sexual arousal (feeling a desire or inclination to engage in sexual activity; Wiemer et al., 2023), during sexual situations are less able to inhibit their subsequent aggressive impulses and more likely to perpetrate sexual assault (Kirwan, 2023; Kirwan & Davis, 2023). This relationship is further exacerbated among men with high trait-level sexual assault predictors, such as impersonal sex (Wilson, 2023) and hostile masculinity (Gildner et al., 2021). Specifically, when experiencing a state negative mood, or a prolonged feeling of negativity which disposes an individual to experiencing more short-term, negative emotions (Prescott-Couch, 2005), some individuals may have difficulties modulating their emotions, making them more likely to use sex to maladaptively regulate their emotions to distract themselves from their negative mood. This may be especially prominent among individuals predisposed to aggression towards women, due to their trait hostile masculinity and impersonal sex, according to previous cross-sectional and longitudinal research (Ellis & Orcutt, 2023; Malamuth et al., 1996). As a result, they may feel more sexually aroused, which could increase their likelihood of perpetrating sexual assault in an attempt to satiate their arousal (Kirwan, 2023; Wiemer et al., 2023).

Similarly, state alcohol intoxication increases state impulsivity and disinhibits sexual assault by causing intoxicated men to focus on cues which impel them towards perpetration, such as sexual arousal (Davis et al., 2021, 2022; George, 2019), instead of cues which would inhibit perpetration if they were sober (Davis, Neilson et al., 2023; George et al., 2024), according to alcohol administration research. For example, alcohol’s cognitive effects may result in men being less able to accurately judge women’s affective and nonaffective cues, causing them to misperceive women as being sexually interested in them, according to experimental research (Treat et al., 2020). Men with high trait hostile masculinity and impersonal sex may be especially susceptible to myopia (Neilson et al., 2023), which results in them being less able to adaptively inhibit their impulses to perpetrate sexual assault when they are intoxicated during sexual situations, according to both cross-sectional and experimental studies (Chen et al., 2024; Davis et al., 2022; Kirwan, Lanni et al., 2019; Kirwan, Davis et al., 2022b). Finally, experiencing a negative mood may result in greater state impulsivity, as individuals attempt to distract themselves from that negative mood (Weiss et al., 2012). This is further exacerbated by trait-level hostile masculinity and impersonal sex, especially during interactions with women, resulting in men who possess these characteristics being more likely to feel impulsive, and to lash out aggressively towards women when experiencing situational factors, including a negative mood, based on cross-sectional and longitudinal research (Malamuth et al., 1996).

The Present Study

In light of previous research and theory, the present study sought to integrate the trait-level sexual assault predictors implicated within the Confluence Model (hostile masculinity and impersonal sex) with state-level predictors consistent with the I3 Model. Specifically, we examined the ways hostile masculinity and impersonal sex interact with induced positive or negative mood and acute alcohol intoxication to inform participants’ state experiences of impulsivity and sexual arousal, state ability to modulate their emotions, and intentions to perpetrate sexual assault during a hypothetical sexual scenario. We anticipated that trait-level Confluence Model impellors (hostile masculinity and impersonal sex; Bruera et al., 2023; Malamuth & Hald, 2017) would predispose men to be more susceptible to established state-level sexual assault impellors which were manipulated in the present study, namely alcohol intoxication and a negative mood (Davis et al., 2021, 2022; George, 2019; Kirwan, 2023), resulting in greater sexual arousal and impulsivity during the hypothetical sexual scenario. In other words, we expected that participants who were intoxicated and in a negative mood would experience greater state sexual arousal and impulsivity if they also reported high levels of hostile masculinity and impersonal sex whereas individuals who were not intoxicated or who were in a positive mood would not show this positive relationship (hypothesis 1). Additionally, we predicted that state sexual arousal and impulsivity would interact such that individuals who were more sexually aroused and more impulsive would have greater state difficulties modulating their emotions during the scenario (hypothesis 2). Finally, state difficulties modulating emotions were expected to be associated with greater intentions to perpetrate sexual assault (hypothesis 3; Figure 1).

Figure 1:

Figure 1:

Open in a new tab

Hypothesized Mode

Methods

Procedures

All materials and procedures for this study were approved by the university’s Institutional Review Board prior to data collection. Participants were recruited from 2019–2023 with online and print advertisements disseminated in an urban region of the southwestern United States for a study which included both an alcohol administration component and daily surveys assessing condom use and sexual activity. The present study represents a secondary data analysis of this research. As such, individuals were eligible to participate if they were 1) male, 2) 21–30 years old, 3) not in a long-term (i.e., six months or more) monogamous relationship, 4) had consumed alcohol at least twice in the past 30 days, 5) had consumed at least three drinks on one occasion within the past six months, 6) had sex with a woman at least twice within the past 30 days and/or intended to have sex with a woman at least twice in the next 30 days, and 7) had condomless sex with a woman at least once in the past year. Also, in line with NIAAA’s guidelines for ethical alcohol administration research (2023), participants who had an alcohol use disorder (Pokorny et al., 1972) or a medical condition/medication which contraindicated alcohol consumption were excluded. Participants in long-term, monogamous relationships and those who were not sexually active were excluded to ensure participants would be able to project themselves into the hypothetical, sexual scenario with a casual sexual partner. People under 21 and who did not regularly consume alcohol were excluded to avoid having participants consume alcohol illegally and to reduce the likelihood that they would have an aversive reaction when consuming alcohol during the experiment. Finally, participants who had not had condomless sex at least once in the past year were excluded because the present study’s primary aims concerned condom use.

Interested individuals completed an online screening survey which assessed their eligibility based on the criteria listed above; those deemed eligible by the screener provided their phone number so that their eligibility could be verified during a phone interview with study staff. Men whose eligibility was verified and who remained interested in participating in the study were scheduled for an in-person laboratory session, during which they completed a background survey in the lab. Afterwards, participants completed online, daily assessments for 32 days, and were scheduled for a second, in-lab session which included additional questionnaires, alcohol administration, a mood induction, and a hypothetical, sexual scenario. Data from the first and second in-lab sessions were used in the present analyses, but data from the daily assessments were not. Data were collected at multiple times to address the study’s primary purposes.

During both in-lab sessions, participants presented an ID to a male research assistant to verify their name and date of birth. Then, participants provided informed consent and completed online surveys privately via Qualtrics. The first session’s questionnaire included assessments of impersonal sex and components of hostile masculinity. The second in-lab session occurred 5–6 weeks after the first session, and involved a two-by-two factorial design to assess the effects of alcohol (sober and intoxicated) and mood (positive and negative) manipulations on participants’ intentions to perpetrate sexual assault during a hypothetical, sexual scenario. At the beginning of the second session, a male research assistant confirmed the participant had not driven himself to the lab, had not consumed alcohol or prescription drugs for the past 24 hours, and had not consumed anything caloric for the past three hours. Next, the research assistant breathalyzed the participant, to ensure that his BrAC was 0.00%, and weighed him to calculate the amount of alcohol needed to bring him to a peak BrAC of .08%. Then, participants completed additional background questionnaires, including an assessment of the third and final component of hostile masculinity (i.e., noncontact harassment), before being asked to complete a mood induction via a hypothetical scenario.

Mood induction.

Participants were randomly assigned to read and project themselves into a scenario describing either a sequence of positive events (e.g., nice weather, finding cash on the sidewalk) to induce a positive mood, or a scenario describing similar, negative events (e.g., bad weather, getting an unexpected bill) to induce a negative mood. Then, participants completed a post-induction manipulation check of their positive and negative emotions.

Alcohol manipulation.

Participants were also randomized to a beverage condition, using block randomization based on their previous sexual assault perpetration. Participants in the alcohol condition were given 0.816 mL ethanol per pound of body weight, administered using 100 proof vodka mixed with cranberry juice in a 1:3 ratio. Participants in the sober condition received an equivalent amount of pure cranberry juice. All participants were accurately informed of whether they would be consuming alcohol or plain juice prior to beverage administration. Drinks were distributed evenly in three cups, and participants were asked to pace their drinking evenly, such that they took three minutes to finish each cup, and nine minutes to finish all three cups. After beverage administration, participants rinsed their mouths to ensure accurate breathalyzer readings, and breathalyzer tests were administered every four minutes until the participants’ BrACs reached a set criterion of 0.05%, to ensure they completed the hypothetical, sexual scenario on the ascending limb of the BAC curve. Each sober participant was yoked to an alcohol participant, to ensure that they received the same number of breathalyzer tests at equivalent intervals.

Hypothetical sexual scenario.

Next, participants projected themselves into a second-person hypothetical, sexual scenario, which occurred on the same day as the events of the mood induction scenario. The scenario paused at key moments to assess participants’ state sexual arousal, impulsivity, difficulties modulating state emotions, and intentions to perpetrate. Participants read about Michele, a casual sexual partner with whom they had previously had both protected and unprotected vaginal sex, and who had just invited the protagonist back to her apartment where they expected to have sex with her again that night. Upon arriving, Michele and the participant engaged in small talk and drank either vodka or soda (corresponding to their beverage condition). Then, they engaged in consensual kissing and foreplay together and Michele expressed interest in having sex until she realized that neither she nor the participant had a condom (Break 1). Michele refused to have sex without a condom, but consented to some additional non-penetrative sexual activities (Break 2). However, when the protagonist tried initiating sexual intercourse with her, she explicitly refused once again. Then, the scenario ended.

Participants

Overall, 14,595 individuals were screened online; 1,271 (8.7%) met the initial eligibility criteria; 677 (52.3%) were verified via screening calls; and 420 (62.0%) participants completed the first in-lab session. Three participants voluntarily withdrew during the daily diary phase; 19 were withdrawn due to a COVID-19 lab closure; and 108 declined to schedule a second in-lab session. Thus, 290 participants (69.1%) completed the second session, but eight of these participants were removed due to missed attention checks (n = 3), incomplete surveys (n = 4), or experimenter error (n = 1). Thus, these analyses consisted of a final sample of 282 participants.

Participants’ average age was 24.6 (SD = 2.8), and most participants identified as White (n = 212, 77.4%). Additionally, 29 participants identified as multiracial (10.6%), ten as Asian American (3.6%), nine as Black or African American (3.3%), three as Native American (1.1%), one as a Pacific Islander (0.4%), ten as some other race (3.6%), and eight declined to identify their race (2.8%). In terms of their ethnicity, 71 participants identified as Hispanic or Latino (25.2%), 209 identified as not Hispanic or Latino (74.1%), and two participants declined to identify their ethnicity (0.7%). Regarding their income level, 37 (13.1%) made less than $21,000 per year; 73 (25.8%) made $21,000-$41,000; 54 (19.1%) made $41,000-$61,000; 111 (39.4%) made more than $61,000 and 7 (2.5%) declined to indicate their current income. All participants were required to have had sexual experiences with women due to the study’s procedures, but 43 participants (15.2%) also reported previous sexual experiences with men.

Measures

Hostile Masculinity.

Participants’ hostile masculinity was assessed using a latent variable comprised of factors demonstrating hypermasculinity through the promotion of hegemonic gender roles, including adversarial heterosexual beliefs, hostility towards women, and noncontact harassment towards women (Reidy et al., 2022). Adversarial heterosexual beliefs (Adversarial Heterosexual Beliefs Scale; Lonsway & Fitzgerald, 1995) and hostility towards women (Hostility Toward Women Scale; Lonsway & Fitzgerald, 1995) were assessed using 15 and ten items, respectively, on seven-point Likert scales from 1 strongly disagree to 7 strongly agree. The mean of each of these scales was computed and entered as an indicator of latent hostile masculinity. Sample items included “I think that most women would lie just to get ahead” from the hostility towards women scale, and “Men and women are generally out to use each other” from the adversarial heterosexual beliefs scale. Participants’ self-reported noncontact sexual harassment towards women (Harassment Questionnaire; Fitzgerald et al., 1999) was assessed with the sum of 21 items asking participants to report the number of times they had engaged in several unwanted or inappropriate sexual behaviors towards women, which did not involve sexual contact. These items were assessed on 11-point scales from 0 never to 10 10 or more, and a sample item was “how many times have you continued to ask a woman for dates, drinks, dinner, etc. even though she said ‘no’”. Hostility towards women and adversarial heterosexual beliefs were assessed during the first lab session, while harassment towards women was assessed during the second lab session, prior to the mood and alcohol manipulations.

Impersonal Sex.

Participants’ impersonal sexual orientation was assessed with a latent variable reflecting their lifetime number of female sex partners across performative oral sex, receptive oral sex, vaginal sex, and anal sex. These four items were assessed from 0 to 101 more than 100 (Davis et al., 2014) during participants’ first lab session, and a sample item was “On how many DIFFERENT WOMEN have you performed oral sex in your lifetime?”

Sexual aggression history.

Participants completed the revised Sexual Experiences Survey (SES; Koss et al., 2007) to assess perpetration of sexual aggression since they were 14 years old during the first lab session to allow for block randomization in the alcohol and mood conditions. The SES assessed participants’ perpetration of nonconsensual sexual contact, verbally coerced sexual contact, attempted rape, and completed rape, using a variety of tactics (e.g. telling lies, threatening to end the relationship, taking advantage of someone who was too intoxicated to consent, threatening or using physical force, etc.). Participants were asked the number of times they had used each tactic to achieve their desired outcome on a four-point scale from 0 0 times to 3 3+ times. Based on past research (Davis et al., 2022) participants were classified into one of three categories based on their responses to the SES: nonperpetrators (those who did not endorse any item on the SES), “moderate” past perpetrators (those who endorsed at least one instance unwanted sexual contact or verbal coercion only), and “severe” past perpetrators (those who endorsed at least one instance of attempted or completed rape).

State emotions.

Participants’ state emotions were assessed using the means of four items each assessed on seven-point scales from 1 not at all to 7 extremely. Participants reported their sexual arousal (“horny”, “aroused”, “turned on”, and “passionate”) and impulsivity (“bold”, “daring”, “reckless”, and “impulsive”) on a seven-point scale from 1 not at all to 7 extremely during Break 1 of the scenario, and their means were included in the hypothesized model. These assessments showed good to excellent reliability (α’s = .84–.95). Participants also rated how emotionally positive and negative they felt from 1 Not at all positive to 7 Extremely positive immediately following the mood induction as manipulation checks.

State difficulties modulating emotions.

Participants’ state difficulties modulating their current emotions during Break 2 of the scenario were assessed using a mean of seven items from the limited ability to modulate current emotional and behavioral responses subscale of the State Difficulties in Emotion Regulation Scale (Lavender et al., 2017), assessed from 1 not at all to 5 completely. A sample item was “My emotions felt out of control”, and this subscale showed good reliability α = .86.

Intentions to perpetrate sexual assault.

Participants’ intentions to perpetrate penetrative sexual assault were assessed using the mean of six items assessing participants’ likelihood of having penetrative sex at the end of the scenario, despite Michele’s explicit non-consent to unprotected sex. Items were each assessed on a seven-point scale from 1 very unlikely to 7 very likely, showed good reliability α = .85, and a sample item was “at this point in the situation, how likely are you to have vaginal sex with Michele without a condom?”

Transparency and Openness

Data and analytical code are available upon reasonable request to the corresponding author. Preliminary data analyses were conducted using SPSS version 29, and power analyses and structural equation modelling was conducted via Mplus version 8.1 (Muthén & Muthén, 2017). This manuscript describes a secondary analysis of previously collected data which was not preregistered.

Results

Preliminary Analyses

Before conducting structural equation modelling (SEM), bivariate correlations, means, and standard deviations for all the study’s variables were assessed (Table 1). Participants’ self-reported state positive and negative emotions following the mood induction were examined as a manipulation check using t-tests. Analyses confirmed that mood condition predicted participants’ positive (t(280) = 43.78, p < .001) and negative emotions (t(280) = −42.35, p < .001) following the mood manipulation, such that participants in the good mood condition experienced more positive emotions (M = 6.62, SD = 0.69) and fewer negative emotions (M = 1.14, SD = 0.41) than participants in the bad mood condition (positive emotions: M = 1.91, SD = 1.07; negative emotions: M = 5.80, SD = 1.24). Additionally, the power of the hypothesized model was assessed using a Monte Carlo simulation. All main effects, two-way interactions, and three-way interactions in the hypothesized model achieved a power of at least .80 recommended by Cohen, based on an alpha level of .05, a sample size of 282, and a medium effect size (f = .30; Cohen, 2013). However, the four-way interactions did not achieve this recommended level of power. Additionally, a confirmatory factor analysis was conducted to establish the construct validity of the hostile masculinity and impersonal sex latent variables. All the latent indicators loaded significantly onto their hypothesized factors, based on the results of a confirmatory factor analysis (Figure 2).

Table 1:

Bivariate correlations, means, and standard deviations of all study variables.

1 2 3 4 5 6 7 8 9 10 11 12 13
1. Hostile Masculinity
2. Adversarial Heterosexual Beliefs .96‡
3. Hostility Towards Women .87‡ .69‡
4. Harassment Towards Women .21‡ .17** .15*
5. Impersonal Sex .12* 0.09 .14* 0.01
6. Pre-manipulation Anger .20‡ .19** .15* .15* −0.01
7. Pre-manipulation Happiness −0.05 −0.02 −0.09 0.00 0.09 −.19**
8. Post-manipulation Anger .14* .17** 0.05 0.10 −0.05 .15* 0.01
9. Post-manipulation Happiness −0.11 −.13* −0.07 −0.01 0.06 −0.06 .27‡ −.77‡
10. Break 1 Impulsivity .22‡ .21‡ .18** .23‡ 0.08 0.11 .14* 0.00 .25‡
11. Break 1 Sexual Arousal .16** .15* .13* .13* 0.07 0.05 .17** 0.03 .21‡ .64‡
12. Break 2 Difficulties Modulating Current Emotions .22‡ .25‡ .13* 0.07 0.03 .16** −0.02 0.06 0.04 .25‡ .27‡
13. End of story Intentions to Perpetrate .19** .21‡ 0.11 .19** 0.02 0.03 0.09 .18** −0.07 .31‡ .25‡ .16**
Mean 0.00 2.07 2.52 11.15 0.00 1.40 4.06 2.71 3.69 3.02 3.73 1.21 2.66
Standard Deviation 0.72 0.83 0.99 16.35 0.86 0.67 1.36 2.01 2.05 1.45 1.62 0.45 1.47
Open in a new tab

Note:

*

p < .05,

**

p < .01,

‡

p < .001

Figure 2:

Figure 2:

Open in a new tab

Confirmatory Factor Analysis of the Hostile Masculinity and Impersonal Sex Latent Variables.

Note: *p < .05, **p < .01, ‡p < .001. Standardized coefficients are shown.

Hostile Masculinity CFA is fully saturated: χ2(0) = 0.00, p < .001, RMSEA = 0.00, CFI = 1.00

Impersonal Sex CFA Global Fit Statistics: χ2(1) = 0.083, p = .77, RMSEA = 0.00, CFI = 1.00

Structural Equation Modelling

The hypothesized model showed poor fit χ2(35) = 171.38, p < .001, RMSEA = .12, CFI = .64. Individual pathways within the model were assessed using a false discovery rate correction (FDR; Benjamini & Hochberg, 1995) to protect against inflated type 1 error. The hypothesized model included non-significant interactions between hostile masculinity, impersonal sex, alcohol condition, and mood condition on both impulsivity (B = .64, SE = .62, p = .31) and sexual arousal (B = −.57, SE = .54, p = .29). Thus, these interactions were trimmed from the model for the purposes of parsimony. However, the hypothesized model results did indicate that the interactions between alcohol condition, mood condition, and impersonal sex (B = −.68, SE = .29, p = .018) and between alcohol condition, mood condition, and hostile masculinity (B = 1.46, SE = .83, p = .08) on sexual arousal approached significance, and so these interactions were retained. Full results of the hypothesized model are provided in Table 2. Finally, Mplus’s model modification indices recommended the addition of a direct path between impersonal sex and intentions to perpetrate; this pathway was consistent with past research (Malamuth et al., 2021), and so it was also added to the final model.

Table 2:

Full results and adjusted alphas of the hypothesized model after applying an FDR correction.

Pathway B SE α-value p-value Reject Null
Hostile Masculinity × Alcohol → Impulsivity −0.04 0.47 .050 .94 No
Alcohol × Mood → Impulsivity −0.03 0.34 .049 .94 No
Hostile Masculinity × Impersonal Sex × Alcohol → Impulsivity −0.04 0.3 .047 .90 No
Impersonal Sex → Sexual Arousal −0.03 0.15 .046 .86 No
Hostile Masculinity × Alcohol × Mood → Impulsivity −0.14 0.72 .044 .84 No
Hostile Masculinity × Impersonal Sex → Impulsivity 0.08 0.23 .043 .74 No
Hostile Masculinity × Mood → Impulsivity 0.07 0.23 .041 .74 No
Impersonal Sex × Alcohol × Mood → Impulsivity −0.12 0.25 .040 .64 No
Impersonal Sex × Mood Condition → Impulsivity −0.1 0.21 .038 .63 No
Hostile Masculinity × Impersonal Sex × Mood → Sexual Arousal −0.27 0.53 .037 .61 No
Impersonal Sex × Alcohol → Impulsivity 0.09 0.16 .035 .57 No
Hostile Masculinity × Alcohol → Sexual Arousal −0.42 0.54 .034 .44 No
Hostile Masculinity × Impersonal Sex × Alcohol × Mood → Sexual Arousal 0.64 0.62 .032 .31 No
Hostile Masculinity × Impersonal Sex × Mood → Impulsivity 0.49 0.46 .031 .29 No
Hostile Masculinity × Impersonal Sex × Alcohol × Mood → Impulsivity −0.57 0.54 .029 .29 No
Hostile Masculinity × Impersonal Sex → Sexual Arousal 0.33 0.27 .028 .21 No
Hostile Masculinity × Mood → Sexual Arousal −1.03 0.66 .026 .12 No
Hostile Masculinity × Impersonal Sex × Alcohol → Sexual Arousal −0.53 0.34 .025 .12 No
Impulsivity → Difficulties Modulating Current Emotions 0.04 0.02 .024 .12 No
Alcohol × Mood → Sexual Arousal −0.62 0.39 .022 .11 No
Hostile Masculinity → Impulsivity 0.58 0.36 .021 .11 No
Impersonal Sex × Alcohol → Sexual Arousal 0.32 0.18 .019 .08 No
Hostile Masculinity × Alcohol × Mood → Sexual Arousal 1.46 0.83 .018 .08 No
Hostile Masculinity → Sexual Arousal 0.76 0.42 .016 .07 No
Impersonal Sex → Impulsivity 0.28 0.13 .015 .04 No
Impulsivity × Sexual Arousal → Difficulties Modulating Current Emotions 0.02 0.01 .013 .03 No
Impersonal Sex × Mood Condition → Sexual Arousal 0.56 0.24 .012 .019 No
Impersonal Sex × Alcohol × Mood → Sexual Arousal −0.68 0.29 .010 .018 No
Mood Condition → Impulsivity 0.64 0.25 .009 .011 No
Alcohol Condition → Impulsivity 0.58 0.22 .007 .010 No
Sexual Arousal → Difficulties Modulating Current Emotions 0.05 0.02 .006 .008 No
Difficulties Modulating Current Emotions → Intentions to Perpetrate 0.59 0.2 .004 .003 Yes
Alcohol Condition → Sexual Arousal 0.8 0.26 .003 .002 Yes
Sexual Arousal ←→ Impulsivity 1.09 0.13 .001 < .001 Yes
Open in a new tab

Note: Unstandardized coefficients are reported. → = Regression, ←→ = Correlation.

The final model fit the data very well χ2(33) = 39.18, p = .21, RMSEA = .03, CFI = .98 (Figure 3). The remaining interactions in the final model were all significant: impersonal sex, alcohol condition, and mood condition interacted to predict sexual arousal (B = −.48, SE = .23, p = .032), such that the positive relationship between impersonal sex and sexual arousal was stronger for those in the sober, positive mood (B = .55, SE = .13, p < .001), intoxicated, positive mood (B = .28, SE = .12, p = .02), and intoxicated, negative mood conditions (B = .23, SE = .10, p = .03), while there was no relationship for sober, negative mood participants (B = .02, SE = .13, p = .87). Hostile masculinity, alcohol condition, and mood condition also interacted to predict sexual arousal (B = 1.50, SE = .65, p = .021) such that there was a positive relationship for those in the sober, negative mood (B = .71, SE = .35, p = .04) and the intoxicated, positive mood conditions (B = .82, SE = .29, p = .01), but not for those in the sober, positive mood (B = −.21, SE = .44, p = .63) and intoxicated, negative mood conditions (B = .24, SE = .28, p = .41; Figure 4). Finally, impulsivity and sexual arousal interacted to predict difficulties modulating emotions during the scenario (B = .02, SE = .01, p = .029), such that sexual arousal was positively related to difficulties modulating emotions among those who reported high impulsivity (B = .07, SE = .03, p = .03), but not among those who reported low impulsivity (B = −.001, SE = .03, p = .96; Figure 5). Greater state difficulties with emotion modulation predicted stronger sexual assault perpetration intentions. Indirect effects of the model’s variables were also examined, though none of the indirect effects were significant. Full results of this final model are provided in Table 3. The final model accounted for 38.6% of the variance in sexual assault perpetration intentions, 10.2% of the variance in state difficulties modulating emotions, 20.3% of the variance in state impulsivity, and 16.1% of the variance in state sexual arousal.

Figure 3:

Figure 3:

Open in a new tab

The final model.

Note: *p < .05, **p < .01, ‡p < .001. Standardized coefficients are shown.

Figure 4:

Figure 4:

Open in a new tab

Hostile masculinity × beverage × mood and impersonal sex × beverage × mood interactions in the final model

Note: Error bars represent 95% confidence intervals.

Figure 5:

Figure 5:

Open in a new tab

Sexual arousal × impulsivity interaction in the final model

Note: Error bars represent 95% confidence intervals.

Table 3:

Full results and adjusted alphas of the final model after applying an FDR correction.

Pathway B SE α-value p-value Reject Null
Impersonal Sex → Sexual Arousal 0.02 0.12 .050 .87 No
Hostile Masculinity × Beverage → Sexual Arousal −0.47 0.44 .048 .29 No
Impersonal Sex × Beverage → Sexual Arousal 0.21 0.15 .045 .16 No
Beverage × Mood → Sexual Arousal −0.45 0.3 .043 .14 No
Impulsivity → Difficulties Modulating Current Emotions 0.04 0.02 .040 .12 No
Hostile Masculinity × Mood → Sexual Arousal −0.92 0.51 .038 .07 No
Hostile Masculinity → Sexual Arousal 0.71 0.36 .036 .05 No
Impersonal Sex × Beverage × Mood → Sexual Arousal −0.48 0.23 .033 .032 Yes
Impulsivity × Sexual Arousal → Difficulties Modulating Current Emotions 0.02 0.01 .031 .029 Yes
Hostile Masculinity × Beverage × Mood → Sexual Arousal 1.5 0.65 .029 .021 Yes
Sexual Arousal → Difficulties Modulating Current Emotions 0.05 0.02 .026 .008 Yes
Beverage Condition → Sexual Arousal 0.67 0.23 .024 .004 Yes
Impersonal Sex × Mood Condition → Sexual Arousal 0.53 0.19 .021 .004 Yes
Difficulties Modulating Current Emotions → Intentions to Perpetrate 0.59 0.2 .019 .003 Yes
Hostile Masculinity → Impulsivity 0.51 0.17 .017 .002 Yes
Beverage Condition → Impulsivity 0.5 0.16 .014 .002 Yes
Sexual Arousal ←→ Impulsivity 1.11 0.13 .012 < .001 Yes
Mood Condition → Sexual Arousal 0.91 0.24 .010 < .001 Yes
Impersonal Sex → Impulsivity 0.27 0.06 .007 < .001 Yes
Mood Condition → Impulsivity 0.65 0.16 .005 < .001 Yes
Impersonal Sex → Intentions to Perpetrate 0.62 0.49 .002 < .001 Yes
Open in a new tab

Note: Unstandardized regression coefficients are reported. → = Regression, ←→ = Correlation.

Discussion

This study addressed the need for research to utilize a more comprehensive ecological framework for sexual assault perpetration. Specifically, we assessed how trait-level Confluence Model factors (i.e., hostile masculinity, impersonal sex) interacted with state and situational I3 factors (e.g., instigation of negative emotional state), to influence men’s emotion modulation and intentions to perpetrate sexual assault. Results showed that interactions between underlying Confluence Model factors and I3 state-specific factors were associated with participants’ in-the-moment experiences of specific emotions, their abilities to modulate those emotions, and intentions to perpetrate sexual assault. However, some of the interaction effects yielded unexpected results.

This study showed limited support for hypothesis 1, and in particular did not provide support for the confluence of impersonal sex and hostile masculinity on sexual arousal or impulsivity. To our knowledge, no research has attempted to examine the confluence of impersonal sex and hostile masculinity on state-level factors such as arousal and impulsivity. The present study’s results suggest that there may be more nuanced paths through which these trait-level factors influence sexual aggression perpetration in the context of intoxication and state-level emotions, especially sexual arousal. Specifically, intoxicated participants in both mood conditions exhibited a positive association between impersonal sex and sexual arousal. Among sober participants, there was also a positive association between impersonal sex and sexual arousal for those in a positive mood, whereas there was no discernible relationship between impersonal sex and sexual arousal for sober individuals in a negative mood. Consistent with hypothesis 1 and alcohol myopia theory, individuals in the intoxicated, negative mood condition may have been more focused on the sexual aspects of the scenario (e.g., kissing and consensual foreplay) than the negative aspects (e.g., having an unpleasant chat with a friend during lunch and getting an unexpected bill in the mail) relative to sober participants in the negative mood condition, resulting in greater arousal (Steele & Josephs, 1990). Inconsistent with hypothesis 1, positive mood may have served as an impellor to participants’ sexual arousal, especially among participants who endorsed impersonal sex, resulting in the unexpected positive association among participants in the positive mood conditions regardless of intoxication.

Results did not support hypothesis 1 with regards to the expected association between hostile masculinity and sexual arousal for intoxicated participants in a negative mood. Instead, there was a positive association between hostile masculinity and sexual arousal for intoxicated participants in a positive mood. This may have indicated that intoxicated participants with greater hostile masculinity were hoping to use sex to maintain their good mood (Jaffe et al., 2023), leading to greater sexual arousal. For sober individuals, those in a negative mood had a positive association between hostile masculinity and sexual arousal. Although this result was unexpected, it is consistent with research noting that men who adhere more strongly to traditional masculine norms often use sex to improve their mood, resulting in strong sexual desires when experiencing negative emotions during sexual situations (Janssen et al., 2008). However, these explanations are speculative, and the studies examining them were qualitative, so future quantitative studies should continue to examine the relationship between hostile masculinity, mood, intoxication, and arousal to confirm that these relationships are not spurious.

Consistent with hypothesis 2, results showed that the positive association between impulsivity and men’s ability to modulate their emotional responses depended on sexual arousal. Specifically, this finding expands on those showing that men who are sexually aroused are more likely to act impulsively to satiate their sexual arousal, including by perpetrating sexual assault (Kirwan, 2023; Wiemer et al., 2023), and suggests that it is the combined effects of state sexual arousal and state impulsivity which impels perpetration through difficulties modulating one’s emotions. Thus, men who experience lower state sexual arousal or lower state impulsivity during sexual situations may experience fewer difficulties modulating their emotions, thereby experiencing less impellance to perpetrate sexual assault.

Hypothesis 3 was supported; the model contained a direct, positive association between difficulties modulating current emotions and intentions to perpetrate. Additionally, we found a direct association between impersonal sex and intentions to perpetrate, which though unanticipated, is not unprecedented in previous research (Malamuth et al., 2021; Wilson, 2023). Due to the limited body of literature attempting to integrate state- and trait-level sexual assault predictors, continued exploration of additional state-level factors, such as masculinity threats (Vescio et al., 2023), that may influence the association between trait-based factors, such as impersonal sex, and sexual assault perpetration is needed.

Strengths

One of this study’s most notable strengths is assessing both trait- and state-factors that predict emotion regulation and sexual assault perpetration. Specifically, we integrated trait factors consistent with the Confluence Model (Malamuth et al., 1991) and state factors consistent with the I3 Model (Finkel et al., 2012) to examine how trait factors may predispose certain men to experience state factors proximally associated with intentions to perpetrate sexual assault. In doing so, this study serves as a first step toward investigating an integrated model that highlights important next steps for future work to continue to examine and refine a model that integrates the Confluence and I3 Theories to disentangle the complex nature of sexual assault perpetration.

Another strength is that over a third of our sample (34.8%) identified as a historically marginalized race or ethnicity. This enhances the study’s generalizability and enriches the validity and applicability of our findings to a broader population. Although additional research should examine the specific characteristics and intersectional considerations associated with these marginalized populations, the present study provides preliminary evidence of the potential contributions of the trait- and state-level factors associated with the Confluence and I3 Models to predict emotion regulation and sexual aggression among diverse populations.

Limitations

The current findings should be interpreted in light of some limitations. First, given our eligibility criteria, the study used a sample comprised of men who reported engaging in non-problematic alcohol use and casual sex with women, with some recent history of condomless sex, and it collected data at multiple time points over a 5–6 week period. Although these criteria were necessary given the primary goals and alcohol administration procedures of the current study (i.e., problematic drinkers had to be excluded for ethical reasons), it is unclear whether our results will generalize to individuals not captured by these inclusion criteria (e.g., women, non-binary individuals, men who have sex with men, individuals who engage in problematic alcohol use or other high risk behaviors, etc.). Second, variables to measure hostile masculinity, impersonal sex, state emotions, state difficulties modulating emotions, and sexual assault perpetration intentions were assessed via self-report, which may have resulted in bias and socially desirable responding. Although relevant procedures were undertaken to maximize participants’ comfort in disclosing these sensitive behaviors during the study (e.g., reassuring participants that their responses would remain confidential, obtaining a Certificate of Confidentiality, completing surveys in private), it remains likely that some participants underreported these sensitive behaviors (Tourangeau & Yan, 2007). Relatedly, our interpretations are limited by the Confluence Model factors included in the current analyses (i.e., hostile masculinity and impersonal sex), and the indicators used to measure these factors (i.e., adversarial heterosexual beliefs, hostility towards women, and harassment towards women and lifetime number of female sex partners across performative oral sex, receptive oral sex, vaginal sex, and anal sex). Thus, it is unclear whether our findings would hold when including additional variables. For example, although it is likely that the number of sexual partners is indicative of an impersonal sex orientation among the current sample of non-monogamous, sexually active men, additional self-report measures of impersonal sex, such as assessing positive attitudes about casual sex and the age at which participants first had sex, would also be useful. Similarly, although hostile masculinity and impersonal sex are considered the core components of the Confluence Model, secondary risk factors such as social desirability, violent pornography use, peer influences, and misperceptions of sexual interest (see Malamuth et al., 2021), could be studied as well. Likewise, the present study may have lacked the statistical power necessary to accurately assess the interaction between hostile masculinity, impersonal sex, mood, and alcohol intoxication. Finally, although hypothetical scenarios are valid analogues for real world sexual assault perpetration (Davis et al., 2014), our assessment of sexual assault perpetration pertained to intentions during a sexual scenario rather than real-world contexts.

Future Directions

Future research is needed to replicate the current findings, for example by cross-validating the results using a separate half sample. Similarly, although the trait-level constructs assessed in the present model tend to be stable, and are capable of predicting behavior over long periods of time (Malamuth et al., 1996), future research in this area should collect data at a single time point. Moreover, future studies may benefit from addressing the aforementioned limitations. For example, studies could examine the associations between proximal and state-level sexual assault predictors among women, couples in committed relationships, and men who have sex with men, as well as among samples which consist primarily or exclusively of more racially and ethnically diverse groups. Similarly, future research should also seek to recruit a larger sample capable of expanding the present study’s model to include additional relevant state- and trait- level predictors, such as self-report measures of impersonal sex (e.g., attitudes towards casual sex), sensation seeking, alcohol expectancies, state anger, pornography use, peer influences, and misperceptions of sexual interest (Kirwan, 2023; Kirwan & Davis, 2023; Kirwan, VanDaalen et al., 2022, 2023; Malamuth et al., 2021; Neilson et al., 2022). This larger sample could also more accurately assess the interaction between hostile masculinity, impersonal sex, mood, and alcohol intoxication. Finally, recruitment for this study occurred in one urban region in the U.S.; thus future research should be conducted in other areas of the U.S. as well as other low-, middle-, and high-income countries, to establish the generalizability of these findings to all these populations.

Clinical Implications

This study’s findings offer preliminary implications for preventing sexual assault perpetration. First, examining the interactions of trait-level hostile masculinity and impersonal sex with state-level alcohol intoxication and mood highlights the complex interplay between such underlying beliefs and in-the-moment or situational factors during sexual situations. Furthermore, the mediating role of state abilities to modulate emotions between state emotions and intentions to perpetrate also underscores the role of emotion regulation in preventing sexual assault. As such, sexual assault prevention interventions should consider both trait- and state-level factors within their programming, such as by targeting men high in trait-level hostile masculinity or impersonal sex. Further, consistent with burgeoning literature on this topic, utilizing mindfulness-based or cognitive-behavioral techniques that provide emotion regulation skills to adaptively enhance men’s capacities to regulate state emotions, such as sexual arousal and impulsivity during sexual situations, may also be useful in preventing sexual assault perpetration (Chen et al., 2024; Davis et al., 2021, 2022; Hammett et al., 2023).

Conclusions

The present study provides an important first step toward integrating the Confluence Model’s trait-level sexual assault predictors with proximal, state-level sexual assault predictors consistent with the I3 Model’s framework. Our findings suggest that alcohol intoxication and affective states may interact with hostile masculinity and impersonal sex tendencies to influence state emotions such as sexual arousal and impulsivity. Experiencing these emotions may make it more difficult for men to modulate their emotions, ultimately increasing the risk of sexual assault perpetration intentions. Future research should further examine these interactions and focus on whether intentions translate into real-world perpetration. Similarly, interventions seeking to reduce sexual assault perpetration may benefit from teaching individuals to use adaptive emotion modulation strategies, especially when intoxicated during sexual situations.

Public Significance Statement.

Underlying (e.g., hostile masculinity and impersonal sex) and in-the-moment (e.g., alcohol intoxication, mood) factors interact to predict men’s sexual arousal during sexual situations. More sexually aroused men struggle to regulate their emotions when they feel more impulsive, and these emotion regulation difficulties are associated with greater intentions to perpetrate sexual assault against women. Findings demonstrate that both underlying and in-the-moment factors are important to consider when designing and evaluating sexual assault prevention programs.

Acknowledgments

Research reported in this publication was supported by the National Institute on Alcohol Abuse and Alcoholism of the National Institutes of Health (NIH) under Award Number R37AA025212 to the last author. The content is solely the responsibility of the authors and does not necessarily represent the official views of NIH. The authors have no conflicts of interest to report.

References

  1. Anderson RE, Silver KE, Ciampaglia AM, Vitale AM, & Delahanty DL (2021). The frequency of sexual perpetration in college men: A systematic review of reported prevalence rates from 2000 to 2017. Trauma, Violence, & Abuse, 22(3), 481–495. DOI: 10.1177/1524838019860619 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological), 57(1), 289–300. DOI: 10.1111/j.2517-6161.1995.tb02031.x [DOI] [Google Scholar]
  3. Bruera JA, Bobbio A, & Arbach K (2023). An examination of the Confluence Model (CM) of risk for sexual aggression. Journal of Sexual Aggression, 29(2), 177–192. DOI: 10.1080/13552600.2022.2071490 [DOI] [Google Scholar]
  4. Chen W, Hammett JF, & Davis KC (2024). Preliminary efficacy of a brief cognitive restructuring intervention to reduce sexual coercion intentions: roles of alcohol and emotion regulation. Journal of Interpersonal Violence, 39(3–4), 910–920. DOI: 10.1177/08862605231197784 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cohen J (2013). Statistical power analysis for the behavioral sciences. Routledge. [Google Scholar]
  6. Davis KC, Kirwan M, Wegner R, Neilson EC, & Stappenbeck CA (2020). Effects of alcohol, condom request style, and state anger on men’s condom use resistance. Journal of Studies on Alcohol and Drugs, 81(4), 454–461. DOI: 10.15288/jsad.2020.81.454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Davis KC, Neilson EC, Kirwan M, Bird ER, Eldridge N, George WH, & Stappenbeck CA (2022). The interplay of sexual arousal and power-related emotions in men’s alcohol-involved sexual aggression intentions. The Journal of Sex Research, 59(6), 765–779. DOI: 10.1080/00224499.2021.1972923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Davis KC, Neilson EC, Kirwan M, Eldridge N, George WH, & Stappenbeck CA (2021). Alcohol-involved sexual aggression: Emotion regulation as a mechanism of behavior change. Health Psychology, 40(12), 940–950. DOI: 10.1037/hea0001048 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Davis KC, Neilson EC, Kirwan M, George WH, & Stappenbeck CA (2023). Understanding Alcohol-Involved Sexual Aggression Through the Science of Behavior Change. In Alcohol and Sexual Violence (pp. 75–103). Cham: Springer International Publishing. DOI: 10.1007/978-3-031-24426-1_4 [DOI] [Google Scholar]
  10. Davis KC, Parrott DJ, George WH, Tharp AT, Hall GCN, & Stappenbeck CA (2014). Studying sexual aggression: A review of the evolution and validity of laboratory paradigms. Psychology of Violence, 4(4), 462–476. DOI: 10.1037/a0037662 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Davis KC, Stappenbeck CA, Norris J, George WH, Jacques-Tiura AJ, Schraufnagel TJ, & Kajumulo KF (2014). Young men’s condom use resistance tactics: A latent profile analysis. The Journal of Sex Research, 51(4), 454–465. DOI: 10.1080/00224499.2013.776660 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Davis KC, Stewart R, Kirwan M, Chen W, & Hammett JF (2023). Coercive condom use resistance: A scoping review. Psychology of Violence, 13(5), 361–373. DOI: 10.1037/vio0000481 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Ellis RA, & Orcutt HK (2023). The indirect effect of avoidant motives for sex on the pathways from childhood maltreatment to risky sex behaviors. Journal of interpersonal violence, 38(21–22), 11337–11355. DOI: 10.1177/08862605231179726 [DOI] [PubMed] [Google Scholar]
  14. Evenden JL (1999). Varieties of impulsivity. Psychopharmacology, 146(4), 348–361. [DOI] [PubMed] [Google Scholar]
  15. Finkel EJ (2014). The I3 model: Metatheory, theory, and evidence. In Advances in experimental social psychology (Vol. 49, pp. 1–104). Academic Press. DOI: 10.1016/B978-0-12-800052-6.00001-9 [DOI] [Google Scholar]
  16. Finkel EJ, DeWall CN, Slotter EB, McNulty JK, Pond RS Jr, & Atkins DC (2012). Using I3 theory to clarify when dispositional aggressiveness predicts intimate partner violence perpetration. Journal of Personality and Social Psychology, 102(3), 533. DOI: 10.1037/a0025651 [DOI] [PubMed] [Google Scholar]
  17. Fitzgerald LF, Magley VJ, Drasgow F, & Waldo CR (1999). Measuring sexual harassment in the military: the sexual experiences questionnaire (SEQ—DoD). Military Psychology, 11(3), 243–263. DOI: 10.1207/s15327876mp1103_3 [DOI] [Google Scholar]
  18. George WH (2019). Alcohol and sexual health behavior: “What we know and how we know it”. The Journal of Sex Research, 56(4–5), 409–424. DOI: 10.1080/00224499.2019.1588213 [DOI] [PubMed] [Google Scholar]
  19. George WH, Blayney JA, & Davis KC (2024). Impact of acute alcohol consumption on sexuality: a look at psychological mechanisms. Annual Review of Clinical Psychology, 20. DOI: 10.1146/annurev-clinpsy-080921-075423 [DOI] [PubMed] [Google Scholar]
  20. Gildner DJ, Kirwan M, Pickett SM, & Parkhill MR (2021). Impulse control difficulties and hostility toward women as predictors of relationship violence perpetration in an undergraduate male sample. Journal of Interpersonal Violence, 36(9–10), NP4653–4678. DOI: 10.1177/0886260518792972 [DOI] [PubMed] [Google Scholar]
  21. Gross JJ (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2(3), 271–299. DOI: 10.1037/1089-2680.2.3.271 [DOI] [Google Scholar]
  22. Hammett JF, Chen W, Stappenbeck CA, & Davis KC (2023). Preliminary efficacy of a cognitive restructuring intervention for intimate partner aggression: emotional clarity as a mechanism of change. Journal of interpersonal violence, 38(9–10), 7037–7046. DOI: 10.1177/08862605221140054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jaffe AE, Blayney JA, Jones HR, Stappenbeck CA, George WH, & Davis KC (2023). Sexual Decision Making When Intoxicated: Women’s Reasons for and Against Having Sex in a Laboratory-Based Scenario. The Journal of Sex Research, 1–16. DOI: 10.1080/00224499.2023.2249774 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Janssen E, McBride KR, Yarber W, Hill BJ, & Butler SM (2008). Factors that influence sexual arousal in men: A focus group study. Archives of Sexual Behavior, 37, 252–265. DOI: 10.1080/00224499.2012.684251 [DOI] [PubMed] [Google Scholar]
  25. Kirwan M (2023). Attitudes, emotions, and emotion regulation as underlying and in-the-moment predictors of men’s intentions to perpetrate sexual assault: An application of the I3 model. The Journal of Sex Research, 1–14. DOI: 10.1080/00224499.2023.2288078 [DOI] [PubMed] [Google Scholar]
  26. Kirwan M, & Davis KC (2023). Profiles of emotion regulation strategies and intentions to perpetrate sexual assault. Psychology of Violence. DOI: 10.1037/vio0000469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kirwan M, Davis KC, Stappenbeck CA, & George WH (2022a). The effect of state, trait, and situational factors on young men’s coercive condom use resistance. Psychology of Men & Masculinities. DOI: 10.1037/men0000399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kirwan M, Davis KC, Stappenbeck CA, & George WH (2022b). The Roles of Emotion Regulation, Alcohol Consumption, and Women’s Condom Request Style in Men’s Coercive Condom Use Resistance Intentions. Journal of aggression, maltreatment & trauma, 31(9), 1187–1205. DOI: 10.1080/10926771.2022.2089862 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kirwan M, Lanni DJ, Warnke A, Pickett SM, & Parkhill MR (2019). Emotion regulation moderates the relationship between alcohol consumption and the perpetration of sexual aggression. Violence Against Women, 25(9), 1053–1073. DOI: 10.1177/1077801218808396 [DOI] [PubMed] [Google Scholar]
  30. Kirwan M, Svenson DW, Pickett SM, & Parkhill MR (2019). Emotion regulation as a mediator between sleep quality and interpersonal aggression. Personality and individual differences, 148, 32–37. DOI: 10.1016/j.paid.2019.05.018 [DOI] [Google Scholar]
  31. Kirwan M, VanDaalen RA, Eldridge N, & Davis KC (2023). Sensation seeking and alcohol expectancies regarding sexual aggression as moderators of the relationship between alcohol use and coercive condom use resistance intentions. Psychology of addictive behaviors, 37(2), 309. DOI: 10.1037/adb0000822 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kirwan M, VanDaalen RA, Eldridge N, & Davis KC (2023). Typical drinking, sensation seeking, and alcohol expectancies regarding sexual coercion as predictors of coercive condom use resistance perpetration. Journal of interpersonal violence, 38(9–10), 7047–7058. DOI: 10.1177/08862605221140045 [DOI] [PubMed] [Google Scholar]
  33. Koss MP, Abbey A, Campbell R, Cook S, Norris J, Testa M, … & White J (2007). Revising the SES: A collaborative process to improve assessment of sexual aggression and victimization. Psychology of Women Quarterly, 31(4), 357–370. DOI: 10.1111/j.1471-6402.2007.00385.x [DOI] [Google Scholar]
  34. Lavender JM, Tull MT, DiLillo D, Messman-Moore T, & Gratz KL (2017). Development and validation of a state-based measure of emotion dysregulation: The State Difficulties in Emotion Regulation Scale (S-DERS). Assessment, 24(2), 197–209. DOI: 10.1177/1073191115601218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lonsway KA, & Fitzgerald LF (1995). Attitudinal antecedents of rape myth acceptance: A theoretical and empirical reexamination. Journal of personality and social psychology, 68(4), 704. DOI: 10.1037/0022-3514.68.4.704 [DOI] [Google Scholar]
  36. Malamuth NM, & Hald GM (2017). The confluence mediational model of sexual aggression. In Boer DP (Ed.), The Wiley Handbook on the Theories, Assessment, and Treatment of Sexual Offending (53–71). John Wiley & Sons, Ltd. DOI: 10.1002/9781118574003.wattso003 [DOI] [Google Scholar]
  37. Malamuth NM, Heavey CL, & Linz D (2014). The confluence model of sexual aggression: Combining hostile masculinity and impersonal sex. In Sex Offender Treatment (pp. 13–38). Routledge. [Google Scholar]
  38. Malamuth NM, Lamade RV, Koss MP, Lopez E, Seaman C, & Prentky R (2021). Factors predictive of sexual violence: Testing the four pillars of the Confluence Model in a large diverse sample of college men. Aggressive behavior, 47(4), 405–420. DOI: 10.1002/ab.21960 [DOI] [PubMed] [Google Scholar]
  39. Malamuth NM, Sockloskie RJ, Koss MP, & Tanaka JS (1991). Characteristics of aggressors against women: testing a model using a national sample of college students. Journal of consulting and clinical psychology, 59(5), 670. DOI: 10.1037/0022-006X.59.5.670 [DOI] [PubMed] [Google Scholar]
  40. Muthén LK and Muthén BO (1998–2017). Mplus User’s Guide. Eighth Edition. Los Angeles, CA: Muthén & Muthén [Google Scholar]
  41. Neilson EC, Smith L, Davis KC, & George WH (2022). Acute alcohol intoxication, state anger, and sexual assault perpetration: The role of state emotion regulation. Psychology of violence, 12(1), 42. DOI: 10.1037/vio0000381 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. NIAAA. (2023). NIAAA guidance for conducting alcohol administration studies with human participants, 2023. National Institute on Alcohol Abuse and Alcoholism. https://www.niaaa.nih.gov/research/guidelines-and-resources/guidance_conducting_alcohol_administration_studies_human_participants [Google Scholar]
  43. Prescott-Couch A (2005). What is a Mood?. The Yale Philosophy Review, 2, 42–58. [Google Scholar]
  44. Pickett SM, Parkhill MR, & Kirwan M (2016). The influence of sexual aggression perpetration history and emotion regulation on men’s aggressive responding following social stress. Psychology of Men & Masculinity, 17(4), 363. DOI: 10.1037/men0000032 [DOI] [Google Scholar]
  45. Pokorny AD, Miller BA, & Kaplan HB (1972). The brief MAST: A shortened version of the Michigan Alcoholism Screening Test. American journal of psychiatry, 129(3), 342–345. DOI: 10.1176/ajp.129.3.342 [DOI] [PubMed] [Google Scholar]
  46. Ray TN, & Parkhill MR (2023). Components of hostile masculinity and their associations with male-perpetrated sexual aggression toward women: A systematic review. Trauma, Violence, & Abuse, 24(2), 355–368. DOI: 10.1177/15248380211030224 [DOI] [PubMed] [Google Scholar]
  47. Reidy DE, Leone RM, Bogen KW, & Swahn MH (2022). The culture of masculinity and sexual violence: Raising boys to be nonviolent men. In Engaging boys and men in sexual assault prevention (pp. 125–148). Academic Press. DOI: 10.1016/B978-0-12-819202-3.00013-4 [DOI] [Google Scholar]
  48. Seto MC (2019). Advancing our scientific understanding of sexual harassment. Archives of sexual behavior, 48(6), 1641–1643. DOI: 10.1007/s10508-019-1426-5 [DOI] [PubMed] [Google Scholar]
  49. Slotter EB, & Finkel EJ (2011). I3 theory: Instigating, impelling, and inhibiting factors in aggression. In Shaver PR & Mikulincer M (Eds.), Human aggression and violence: Causes, manifestations, and consequences (pp. 35–52). American Psychological Association. DOI: 10.1037/12346-002 [DOI] [Google Scholar]
  50. Steele CM, & Josephs RA (1990). Alcohol myopia: Its prized and dangerous effects. American psychologist, 45(8), 921. DOI: 10.1037/0003-066X.45.8.921 [DOI] [PubMed] [Google Scholar]
  51. Tourangeau R, & Yan T (2007). Sensitive questions in surveys. Psychological bulletin, 133(5), 859. DOI: 10.1037/0033-2909.133.5.859 [DOI] [PubMed] [Google Scholar]
  52. Treat TA, McMurray B, Betty JR, & Viken RJ (2020). Tracking men’s perceptions of women’s sexual interest. Current Directions in Psychological Science, 29(1), 71–79. DOI: 10.1177/0963721419884322 [DOI] [Google Scholar]
  53. Trottier D, Nolet K, Benbouriche M, Bonneville V, Racine-Latulippe F, & Bergeron S (2021). Sexual violence perpetration and victimization: Providing prevalence rates for understudied populations. Violence and Gender, 8(2), 59–66. DOI: 10.1089/vio.2020.003 [DOI] [Google Scholar]
  54. Vescio TK, Schermerhorn NE, Lewis KA, Yamaguchi-Pedroza K, & Loviscky AJ (2023). Masculinity threats sequentially arouse public discomfort, anger, and positive attitudes toward sexual violence. Personality and social psychology bulletin. DOI: 10.1177/01461672231179431 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Weiss NH, Tull MT, Viana AG, Anestis MD, & Gratz KL (2012). Impulsive behaviors as an emotion regulation strategy: Examining associations between PTSD, emotion dysregulation, and impulsive behaviors among substance dependent inpatients. Journal of anxiety disorders, 26(3), 453–458. DOI: 10.1016/j.janxdis.2012.01.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wiemer J, Kurstak S, Sellmann F, & Lindner K (2023). Sexual stimuli cause behavioral disinhibition in both men and women, but even more so in men. Archives of sexual behavior, 1–16. DOI: 10.1007/s10508-022-02514-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wilson TP (2023). Examining the Role of Alcohol Use and Self-regulation Processes in an Expansion of the Confluence Model of Sexual Aggression (Doctoral dissertation, The University of Texas at San Antonio; ). [Google Scholar]

RESOURCES